Abstract
Purpose :
Primary open angle glaucoma (POAG) is a genetically and clinically heterogeneous disease characterized by retinal ganglion cell apoptosis. In neurodegenerative diseases of the brain, including Alzheimer’s disease (AD), neuronal death leads to induction of a neurodegeneration-associated transcriptional phenotype in microglia, the resident macrophages of the brain. Here we sought to identify genetic associations between POAG and APOE and TREM2, two genes that are known to be critically involved in regulation of microglial neurodegenerative phenotype and are established risk factors for AD.
Methods :
APOE genotypes (alleles 2, 3, and 4) were identified using imputed data from the NEIGHBOR consortium (2606 cases and 2606 controls). The dataset was analyzed for association with POAG and the high-tension (HTG) and normal-tension (NTG) subgroups using logistic regression controlling for age, gender and population-specific eigenvectors and covariates. A likelihood ratio test was done to compare the logistic regression models. TREM2 data was extracted from exome-wide genotypes using the human exome bead array.
Results :
Significant association was observed for the APOE4 allele in POAG overall (P= 0.0022, OR 0.834) and in both the HTG and NTG subgroups (P=0.0052, OR 0.801) and (P= 0.0014, OR 0.712) respectively. Significant association was not observed for either the APOE2 or APOE3 alleles. Interestingly, the APOE4 allele effect is protective in POAG and in HTG and NTG, with greatest effect in NTG. The likelihood ratio test (LRT) confirmed these findings and demonstrated the largest effect with NTG (P= 0.0041). A rare TREM2 variant (A105V) was found only in POAG cases (3 of 2846 cases, allele frequency 0.00053) and in none of 2606 controls. Compared with population data from ExAC the distribution in cases was significant (P=0.02). Other TREM2 rare variants associated with AD, including R62H, were not significantly associated with POAG (P > 0.24).
Conclusions :
We have found that the APOE4 allele is protective in POAG, especially NTG, in this relatively large dataset. Interestingly, the risk effect is similar to what has been observed for age-related macular degeneration but opposite of that for AD, suggesting a mechanistic difference between neurodegenerative diseases of the eye and the brain. Rare TREM2 variants may also contribute to POAG risk, however these appear to be different variants when compared with AD.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.